Re: [PATCH v2 4/8] vsock/virtio: Resize receive buffers so that each SKB fits in a page

[David Laight <david.laight.linux@gmail.com>]

On Tue,  1 Jul 2025 17:45:03 +0100
Will Deacon <will@kernel.org> wrote:

> When allocating receive buffers for the vsock virtio RX virtqueue, an
> SKB is allocated with a 4140 data payload (the 44-byte packet header +
> VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE). Even when factoring in the SKB
> overhead, the resulting 8KiB allocation thanks to the rounding in
> kmalloc_reserve() is wasteful (~3700 unusable bytes) and results in a
> higher-order page allocation for the sake of a few hundred bytes of
> packet data.
> 
> Limit the vsock virtio RX buffers to a page per SKB, resulting in much
> better memory utilisation and removing the need to allocate higher-order
> pages entirely.
> 
> Signed-off-by: Will Deacon <will@kernel.org>
> ---
>  include/linux/virtio_vsock.h     | 1 -
>  net/vmw_vsock/virtio_transport.c | 7 ++++++-
>  2 files changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
> index eb6980aa19fd..1b5731186095 100644
> --- a/include/linux/virtio_vsock.h
> +++ b/include/linux/virtio_vsock.h
> @@ -109,7 +109,6 @@ static inline size_t virtio_vsock_skb_len(struct sk_buff *skb)
>  	return (size_t)(skb_end_pointer(skb) - skb->head);
>  }
>  
> -#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE	(1024 * 4)
>  #define VIRTIO_VSOCK_MAX_BUF_SIZE		0xFFFFFFFFUL
>  #define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE		(1024 * 64)
>  
> diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
> index 488e6ddc6ffa..3daba06ed499 100644
> --- a/net/vmw_vsock/virtio_transport.c
> +++ b/net/vmw_vsock/virtio_transport.c
> @@ -307,7 +307,12 @@ virtio_transport_cancel_pkt(struct vsock_sock *vsk)
>  
>  static void virtio_vsock_rx_fill(struct virtio_vsock *vsock)
>  {
> -	int total_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE + VIRTIO_VSOCK_SKB_HEADROOM;
> +	/* Dimension the SKB so that the entire thing fits exactly into
> +	 * a single page. This avoids wasting memory due to alloc_skb()
> +	 * rounding up to the next page order and also means that we
> +	 * don't leave higher-order pages sitting around in the RX queue.
> +	 */
> +	int total_len = SKB_WITH_OVERHEAD(PAGE_SIZE);

Should that be an explicit 4096?
Otherwise it is very wasteful of memory on systems with large pages.

	David

>  	struct scatterlist pkt, *p;
>  	struct virtqueue *vq;
>  	struct sk_buff *skb;